Excess deposition and accumulation of extracellular matrix (ECM) is found in diseases such as fibrosis of the kidney or lung. Although the cytokine transforming growth factor Beta (TGFβ) regulates extracellular matrix deposition for tissue repair, overproduction of TGFβ clearly underlies tissue fibrosis caused by excess deposition of extracellular matrix resulting in disease (Border and Ruoslahti, J. Clin. Invest. 90:1-7 (1992)). TGFβ's fibrogenic action results from simultaneous stimulation of matrix protein synthesis (Border et al., Kidney Int 37:689-695 (1990), inhibition of matrix degradation and turnover and enhanced cell-matrix interactions through modulation of integrin receptors that facilitate ECM assembly. Overproduction of TGFβ has been demonstrated in glomerulonephritis (Okuda et al., J. Clin. Invest. 86:453-462 (1990)), diabetic nephropathy and hypertensive glomerular injury and in related fibrotic disorders of the lung, liver, heart, arterial wall, skin, brain, joints and bone marrow (Border and Noble, N. Eng. J. Med. 331:128-1292 (1994)). In addition to the kidney, blocking the action of TGFβ with an agent such as antibody or the proteoglycan decorin has been shown to be therapeutic in fibrosis and scarring of the skin, lung, central nervous system and arterial wall (Border and Noble, Kidney Int. 51:1388-1396 (1997)).
Suppression of the production of ECM and prevention of excess accumulation of mesangial matrix in glomeruli of glomerulonephritic rats has been demonstrated by intravenous administration of neutralizing antibodies specific for TGFβ (Border et al., Nature 346:371-374 (1990)) or administration of purified decorin, a proteoglycan (Border et al., Nature 360:361-364 (1992)) and by introduction of nucleic acid encoding decorin, a TGFβ-inhibitory agent, into a rat model of acute mesangial glomerulonephritis (Isaka et al., Nature Med. 2:418-423 (1996)). Inhibition of TGFβ activity, using for example anti-TGFβ antibodies, has been shown to to disrupt TGFβ overproduction (Sharma et al., Diabetes 45:522-530 (1996)).
Dermal scarring following dermal injury results from excessive accumulation of fibrous tissue made up of collagen, fibronectin and proteoglycans at a wound site. Because the fibrous extracellular matrix lacks elasticity, scar tissue can impair essential tissue function as well as result in an undesirable cosmetic appearance. TGFβ is believed to induce the deposition of fibrous matrix at the wound site (Shah et al., Lancet 339:213-214 (1992)).
One explanation for persistent TGFβ overexpression in progressive fibrotic kidney disease is that repeated or multiple episodes of tissue injury, such as occurs in chronic diseases such as hypertension, diabetes or immune complex disease lead to continuous overproduction of TGFβ and extracellular matrix resulting in tissue fibrosis (See Border and Noble, N. Eng. J. Med. 331:1286-1292(1994)). Another possible explanation for persistent TGFα overexpression is the presence of a biologically complex interconnection between TGFβ and the renin-angiotensin system (RAS) in the kidney as part of an emergency system that responds to the threat of tissue injury as discussed further herein.
Renin is an aspartyl proteinase synthesized by juxtaglomerular kidney cells and mesangial cells in humans and rats. (Chansel et al., Am. J. Physiol. 252:F32-F38 (1987) and Dzau and Kreisberg, J. Cardiovasc. Pharmacol. 8(Suppl 10):S6-S10(1986)). Renin plays a key role in the regulation of blood pressure and salt balance. Its major source in humans is the kidney where it is initially produced as preprorenin. Signal peptide processing and glycosylation are followed by secretion of prorenin and its enzymatically active form, mature renin. The active enzyme triggers a proteolytic cascade by cleaving angiotensinogen to generate angiotensin I, which is in turn converted to the vasoactive hormone angiotensin II by angiotensin converting enzyme (“ACE”).
The sequence of the human renin gene is known (GenBank entry M2690). Recombinant human renin has been synthesized and expressed in various expression systems (Sielecki et al., Science 243:1346-1351(1988), Mathews et al., Protein Expression and Purification 7:81-91 (1996)). Inhibitors of renin's enzymatic site are known (Rahuel et al., J. Struct. Biol. 107:227-236 (1991); Badasso et al., J. Mol. Biol. 223:447-453 (1992); and Dhanaraj et al., Nature 357:466-472 (1992)) including an orally active renin inhibitor in primates, Ro 42-5892 (Fischli et al., Hypertension 18:22-31(1991)). Renin-binding proteins and a cell surface renin receptor on human mesangial cells have been identified (Campbell and Valentijn, J. Hypertens. 12:879-890 (1994), Nguyen et al., Kidney Internat. 50:1897-1903 (1996) and Sealey et al., Amer. J. Hyper. 9:491-502 (1996)).
The renin-angiotensin system (RAS) is a prototypical systemic endocrine network whose actions in the kidney and adrenal glands regulate blood pressure, intravascular volume and electrolyte balance. In contrast, TGFβ is considered to be a prototypical cytokine, a peptide signaling molecule whose multiple actions on cells are mediated in a local or paracrine manner. Recent data however, indicate that there is an intact RAS in many tissues whose actions are entirely paracrine and TGFα has wide-ranging systemic (endocrine) effects. Moreover, RAS and TGFε act at various points to regulate the actions of one another.
In a systemic response to an injury such as a wound, the RAS rapidly generates AII that acts by vasoconstriction to maintain blood pressure and later stimulates the secretion of aldosterone, resulting in an increase in intravascular volume. In the wound, TGFβ is rapidly released by degranulating platelets and causes a number of effects including: 1) autoinduction of the production of TGFβ by local cells to amplify biological effects; 2) chemoattraction of monocyte/macrophages that debride and sterilize the wound and fibroblasts that begin synthesis of ECM; 3) causing deposition of new ECM by simultaneously stimulating the synthesis of new ECM, inhibiting the proteases that degrade matrix and modulating the numbers of integrin receptors to facilitate cell adhesion to the newly assembled matrix; 4) suppressing the proinflammatory effects of interleukin-1 and tumor necrosis factor; 5) regulating the action of platelet derived growth factor and fibroblast growth factor so that cell proliferation and angiogenesis are coordinated with matrix deposition; and 6) terminating the process when repair is complete and the wound is closed (Border and Noble, Scientific Amer. Sci. & Med. 2:68-77 (1995)).
Interactions between RAS and TGFβ occur at both the systemic and molecular level. It has been shown that TGFβ's action in causing ECM deposition in a healing wound is the same action that makes TGFβ a powerful fibrogenic cytokine. (Border and Noble, New Engl. J. Med. 331:1286-1292(1994); and Border and Ruoslahti, J. Clin. Invest. 90:107 (1992). Indeed, it is the failure to terminate the production of TGFβ that distinguishes normal tissue repair from fibrotic disease. RAS and TGFβ co-regulate each other's expression. Thus, both systems may remain active long after an emergency response has been terminated, which can lead to progressive fibrosis. The kidney is particularly susceptible to overexpression of TGFβ. The interrelationship of RAS and TGFβ may explain the susceptibility of the kidney to TGFβ overexpression and why pharmacologic suppression of RAS or inhibition of TGFβ are both therapeutic in fibrotic diseases of the kidney. (Noble and Border, Sem. Nephrol., supra and Border and Noble, Kidney Int. 51:1388-1396 (1997)).
Activation of RAS and generation of angiotensin II (AII) are known to play a role in the pathogenesis of hypertension and renal and cardiac fibrosis. TGFβ has been shown to be a powerful fibrogenic cytokine, acting simultaneously to stimulate the synthesis of ECM, inhibit the action of proteases that degrade ECM and increasing the expression of cell surface integrins that interact with matrix components. Through these effects, TGFβ rapidly causes the deposition of excess ECM. AII infusion strongly stimulates the production and activation of TGFβ in the kidney. (Kagami et al., J. Clin. Invest. 93:2431-2437 (1994)). Angiotensin II also upregulates TGFβ production and increases activation when added to cultured vascular smooth muscle cells (Gibbons et al, J. Clin. Invest. 90:456-461 (1992)) and this increase is independent of pressure (Kagami et al., supra). AII also upregulates TGFβ receptors, even in the presence of exogenously added TGFβ which normally down-regulates its own receptors, leading to enhanced TGFβ signalling and enhanced fibronectin production (Kanai et al., J. Am. Soc. Nephrol. 8:518A (1997)). Blockade of AII reduces TGFβ overexpression in kidney and heart, and it is thought that TGFβ mediates renal and cardiac fibrosis associated with activation of RAS (Noble and Border, Sem. Nephrol. 17(5):455-466 (1997)), Peters et al., Kidney International 54 (1998)). Blockade of AII using inhibitors of ACE slow the progression of renal fibrotic disease (see, e.g., Anderson et al., J. Clin. Invest. 76:612-619 (1985) and Noble and Border, Sem. Nephrol. 17(5):455-466 (1997)). What is not clear is whether angiotensin blockade reduces fibrosis solely through controlling glomerular hypertension and thereby glomerular injury, or whether pressure-independent as well as pressure-dependent mechanisms are operating. While ACE inhibitors and AII receptor antagonists have been shown to slow the progress of fibrotic diseases, they do not halt disease and TGFβ levels remain somewhat elevated. (Peters et al., supra).
Thus, RAS and TGFβ can be viewed as powerful effector molecules that interact to preserve systemic and tissue homeostasis. The response to an emergency such as tissue injury is that RAS and TGFβ become activated. Continued activation may result in chronic hypertension and progressive tissue fibrosis leading to organ failure. Because of the interplay between the RAS and TGFβ, and the effects of this interplay on tissue homeostasis, blockade of the RAS may be suboptimal to prevent or treat progressive fibrotic diseases such as diabetic nephropathy.
Components of the renin-angiotensin system act to further stimulate production of TGFβ and plasminogen activator inhibitor leading to rapid ECM accumulation. The protective effect of inhibition of the renin-angiotensin system in experimental and human kidney diseases correlates with the suppression of TGFβ production. (Noble and Border, Sem. Nephrol., supra; and Peters et al., supra).
The renin molecule has been shown to enzymatically cleave angiotensinogen into angiotensin I. The angiotensin I is then converted by Angiotensin Converting Enzyme (“ACE”) to Angiotensin II which acts as an active metabolite and induces TGFβ production. Angiotensin II is an important modulator of systemic blood pressure. It has been thought that if you decrease hypertension by blocking AII's vasoconstrictor effects fibrotic disease is reduced.
In the glomerular endothelium, activation of RAS and TGFβ have been shown to play a role in the pathogenesis of glomerulonephritis and hypertensive injury. Volume (water) depletion and restriction of potassium have been shown to stimulate both production of renin and TGFβ in the juxtaglomerular apparatus (JGA) of the kidney (Horikoshi et al., J. Clin. Invest. 88:2117-2122 (1992) and Ray et al., Kidney Int. 44:1006-1013 (1993)). Angiotensin blockade has also been shown to increase the production of renin. TGFβ has been shown to stimulate the release of renin from kidney cortical slices and cultured JG cells (Antonipillai et al., Am. J. Physiol. 265:F537-F541 (1993); Ray et al., Contrib. Nephrol. 118:238-248 (1996) and Veniant et al., J. Clin. Invest. 98:1996-19970 (1996)), suggesting that renin and TGFβ are coregulated. Other interactions between RAS and TGFβ include that AII induces the production of TGFβ in cultured cells and in vivo (Kagami et al., supra) and AII regulates expression of TGFβ receptors (Kanai et al., 1977, supra). It is thus likely that the fibrogenic effects that have been attributed to AII are actually mediated by TGF.
Another interplay between RAS and TGFβ is with the production of aldosterone. Aldosterone overproduction has been linked to hypertension and glomerulosclerosis. AII stimulates the production and release of aldosterone from the adrenal gland. In contrast, TGFβ suppresses aldosterone production and blocks the ability of AII to stimulate aldosterone by reducing the number of AII receptors expressed in the adrenal (Gupta et al., Endocrinol. 131:631-636 (1992)), and blocks the effects of aldosterone on sodium reabsorption in cultured renal collecting duct cells (Husted et al., Am. J. Physiol. Renal, Fluid Electrolyte Physiol. 267:F767-F775 (1994)). Aldosterone may have fibrogenic effects independent of AII, and may upregulate TGFβ expression. The mechanism of aldosterone's pathological effects is unknown but might be due to stimulation of TGFβ production in the kidney (Greene et al., J. Clin. Invest. 98:1063-1068 (1996)).
Prorenin or renin may have AII-independent actions to increase fibrotic disease. Prorenin overexpressing rats were found to be normotensive but to develop severe glomerulosclerosis (Veniant et al., J. Clin. Invest. 98:1996-1970 (1996)).
Human recombinant renin added to human mesangial cells induces marked upregulation of production of plasminogen activator inhibitors (e.g. PAI-1 and PAI-2) which block the generation of plasmin, a fibrinolytic enzyme important in the dissolution of clots after wounding generated from plasminogen by two enzymes called plasminogen activators, urokinase (u-PA) and tissue plasminogen activator (t-PA). PAI-1 and 2 regulate U-PA and t-PA in turn. Plasmin appears to be a key mediator of extracellular matrix degradation, carrying out at least three functions important to extracellular matrix degradation. Plasmin directly degrades proteoglycan components of extracellular matrix, proteolytically activates metalloproteinases (MMPs) that, in turn, degrade collagens and other matrix proteins, and enzymatically inactivates tissue inhibitors of MMPs (TIMPs), releasing MMPs from inhibition of TIMPs, allowing them to proteolytically digest matrix proteins. (Baricos et al., Kidney Int'l. 47:1039-1047(1995); Baricos et al., J. Amer. Soc. Nephrol. 10:790-795 (1999)). The net generation of active plasmin from the inactive precursor plasminogen results from a balance of the plasminogen activators and PAI-1 and 2, and other factors. PAI-1 binds to vitronectin. (Lawrence et al., J. Biol. Chem. 272:7676-7680 (1997)). Mutant PAI-1 molecules have been developed that have enhanced properties for PAI-1 binding to vitronectin molecules, but do not inhibit either t-PA or u-PA activity, resulting in an increase in the amount of the active form of plasmin. (See, WO 97/39028, Lawrence et al.). PAI-1 is increased in response to added TGFβ (Tomooka et al., Kidney Int. 42:1462-1469 (1992)).
It has been suggested that TGFβ enhances release of renin from storage granules in the juxtaglomerular apparatus of the kidney (Antonipillai et al., Am. J. Physiol. 265:F537-F541 (1993) and Ray et al., Contrib. Nephrol. 118:238-248 (1996)).
Thus, the interactions of RAS and TGFβ production form a complex system which impacts fibrotic ECM accumulation and the incidence of fibrotic disease. Various RAS components such as aldosterone, prorenin and renin may be connected with TGFβ production and fibrotic ECM accumulation. Any successful therapeutic regime must take into account these complex relationships to optimize inhibition of TGFβ to prevent and/or reduce ECM accumulation.
The multiple pathways resulting in TGFβ overexpression and fibrosis proposed from in vitro studies are depicted in FIG. 1. (See, Kagami et al., J. Clin. Invest. 93:2431-2437 (1994); Gibbons et al., J. Clin. Invest. 90:456-461 (1992); Abboud, Kidney Int. 41:581-583 (1992); Ruiz-Ortega et al., J. Am. Soc. Nephrol. 5:683 (1994) abstract; Kim et al., J. Biol. Chem. 267:13702-13707 (1992); Ohno et al., J. Clin. Invest. 95:1363-1369 (1995); Riser et al, J. Clin. Invest. 90:1932-1943 (1992); Riser et al., J. Am. Soc. Nephrol. 4:663 (1993); Ziyadeh et al., J. Clin. Invest. 93:536-542 (1994); Rocco et al., Kidney Int. 41:107-114 (1992); Flaumenhaft et al., Advan. Pharmacol. 24:51-76 (1993); Lopez-Armanda et al., J. Am. Soc. Nepbrol. 5:812 (1994) abstract; Sahai et al., J. Am. Soc. Nephrol. 6:910 (1995); Remuzzi et al., Kidney Int. 1:2-15 (1997); and Remuzzi et al., J. Am. Soc. Nephrol. 9:1321-1332 (1998)). This diagram shows that a large number of factors implicated in kidney injury are believed to increase the production of TGFβ.
In fibrotic diseases overproduction of TGFβ results in excess accumulation of extracellular matrix which leads to tissue fibrosis and eventually organ failure. Accumulation of mesangial matrix is a histological indication of progressive glomerular diseases that lead to glomerulosclerosis and end-stage kidney disease (Klahr et al., N. Engl. J. Med. 318:1657-1666 (1988); Kashgarian and Sterzel, Kidney Int. 41:524-529 (1992)). Rats injected with antithymocyte serum are an accepted model of human glomerulonephritis and this model has demonstrated that overproduction of glomerular TGFβ can underlie the development of glomerulosclerosis (Okuda et al., J. Clin. Invest. 86:453-462 (1990); Border et al., Nature (Lond.) 346:371-374 (1990); Kagami et al., Lab. Invest. 69:68-76 (1993); and Isaka et al., J. Clin. Invest. 92:2597-2602 (1993)). Using cultured rat mesangial cells where the effects of Angiotensin II on glomerular pressure are not a factor, Angiotensin II has been shown to induce TGFβ production and secretion by mesangial cells, and this in turn has been shown to stimulate extracellular matrix production and deposition (Kagami et al., J. Clin. Invest. 93:2431-2437 (1994)). Increases in PAI-1 levels result in decreased degradation of extracellular matrix (Baricos et al., Kidney Int. 47:1039-1047 (1995)). Increases in TGFβ result in increased PAI-1 levels (Tomooka et al., Kidney Int. 42:1462-1469 (1992)). It has been demonstrated that decreasing TGFβ overexpression in a rat model of glomerulonephritis by in vivo injection of neutralizing antibodies to TGFβ, reduces TGFβ overexpression (Border et al., Nature 346:371-374 (1990)), and reduces PAI-1 deposition into the pathological matrix (Tomooka et al., Kidney Int. 42:1462-1469 (1992)). Therefore, decreases in TGFβ levels should result in decreased PAI-1 levels and increased degradation of extracellular matrix to ameliorate organ impairment and fibrotic disease. However, patients present with fibrotic disease that is well advanced in terms of build-up of extra-cellular matrix (ECM). This is because abnormal organ function is undetectable until ECM accumulation is very advanced. For example, in the kidney, standard diagnostic tests do not provide an abnormal reading until about fifty percent of organ function has been lost.
The treatment of conditions associated with excess accumulation of ECM has also focused on decreasing stimuli to disease such as to lower blood pressure or, in the case of diabetic nephropathy to reduce plasma glucose levels. For example, current therapies for treating fibrotic disease in the kidney are limited to AII blockade using ACE inhibitors such as Enalapril or AII receptor antagonists such as Losartan. In addition, patients are encouraged to follow low protein diets since this regimen has some therapeutic value (Rosenberg et al., J. Clin. Invest. 85:1144-1149(1992)). These therapies, at best, prolong organ function by only 1-2 years. This may be because of the multiple pathways that result in TGFβ overexpression or enhanced activity. Moreover, it is likely that current therapeutic strategies to reduce TGFβ overproduction may lead to upregulation of other pathways resulting in continued TGFβ overproduction. For example, when the action of AII is blocked, renin is upregulated which itself increases TGFβ production (see co-pending U.S. patent application, U.S. Ser. No. 09/005,255, incorporated in its entirety herein). More recently, treatments aimed to halt the overproduction of TGFβ have been proposed (Border and Noble, Kidney Internatl. 54 (1998); and Peters et al., Kidney Internatl. 54 (1998)).
Therefore, the most promising therapeutic methods will need to increase ECM degradation to restore organ function as well as decrease TGFβ overproduction and/or activity. Enhanced degradation of excess accumulated ECM can be used to optimize overall reduction in levels of accumulated ECM to restore function to tissues and organs. Proteases that are able to degrade ECM are known. For example, the serine protease plasmin degrades ECM proteins and activates pro-metalloproteinases, in addition to degrading fibrin (Baricos et al., supra). One goal of therapeutic intervention to increase ECM degradation for treating fibrosis could be increasing plasmin in the region of excess ECM deposition.
There is a need for improved therapies to normalize TGFβ production, that take into account the multiple pathways that stimulate TGFβ production, to prevent or reduce excess accumulation of ECM, to restore function to tissues and organs in which excess ECM has accumulated and/or to reduce scar formation at a wound site.